J . gen. Mimobiol. (1962), 29, 719-730
719
Printed in Great Britain
Effect of Chilling on Aerobacter aerogenes in
Aqueous Suspension
BY R. E. STRANGE AND F . A. DARK
Microbiological Research Establishment, Porton, near Salisbury, Wilts.
(Received 13 April 1962)
SUMMARY
Tilelethal effectof cold shockon Aerobacter aerogenes suspensionsdepended
on the time of exposure to low temperature, the growth phase, the concentration of bacteria, the diluent. No death occurred when weak suspensions of
susceptible bacteria (about 108/ml.) in buffered saline (pH 6-5) were
rapidly cooled to 0' and immediately warmed to zoo, but loss of viability
was progressive during 1 hr. at 0'. Bacteria harvested from defined
medium a t intervals during the exponential growth phase varied in
sensitivity to chilling but were more susceptible than stationary phase
organisms. While growing in partially synchronized culture the sensitivity
of bacteria did not increase significantly during the division lag phase. The
viability of dense suspensions (about 1010 bacteria/ml.) in buffered saline
was little affected by chilling for 1hr. a t O", irrespective of the growth
phase. A bacteria-free filtrate from a chilled concentrated suspension of
exponential-phase organisms substantially protected a dilute suspension
from the lethal effect of chilling. Substances found in protective filtrates
were amino acids, adenosine triphosphate and nucleic acid constituents.
When added to the diluent in which susceptible bacteria were chilled, a
mixture of amino acids afforded some protection; small amounts of
adenosine triphosphate had no effect. Other substances found to protect
susceptible bacteria were sucrose (0.3M), magnesium or calcium ions
(5 x 1 0 - 3 ~ ) and, to a much smaller extent, spermine ( W - ~ M ) . The present
results support the suggestion that the lethal effect of chilling is at least
partly due to interference with the functioning of a bacterial permeability
control mechanism.
INTRODUCTION
Sudden chilling causes loss of viability of suspensions of exponential phase
Escherichia coli (Sherman & Albus, 1923; Sherman & Cameron, 1934; Hegarty &
Weeks, 1940; Meynell, 1958) and Pseudomonas pyocyanea (Gorrill & McNeil, 1960).
Meynell(l958) showed that the lethal effect was not due to sudden cooling in itself
by demonstrating that survival was complete either after gradual cooling in a
potentially lethal diluent or after sudden chilling in a solution of sucrose ( 0 . 3 ~ ) .
He suggested the lethal effect may be due to interference with an adaptive mechanism which prevents entry of water into the organism and that the mechanism has
negligible activity at 4". Bacterial death following chilling was not accompanied by
lysis and Meynell detected no differences between electron micrographs of chilled
and unchilled bacteria from the same culture. Gorrill & McNeil (1960) could not
distinguish morphological differences between cold shocked and unshocked P. pyo-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
720
R. E. STRANGE
AND F. A. DARK
cyanea with either dark-ground, phase-contrast or ordinary light microscopy. The
present paper records a study of cold shock in Aerobacter aerogenes.
METHODS
Aerobacter aerogenes strain NCTC 418 was obtained from Professor Sir Cyril
Hinshelwood's laboratory.
Medium and cultural conditions. Organisms were grown at 37" and a t a continuously maintained pH value of 7.2-7-4 with adequate aeration in a batch culture
vessel containing the defined carbon-limitingmedium previously described (Strange,
Dark & Ness, 1961). Partially synchronized cultures were obtained by depriving
organisms of an energy source for a period before growth (McNair Scott & Chu,
1958): medium was seeded with a suspension of washed stationary-phase bacteria
which had been held in buffered saline (pH 6.5) for 20 hr. a t 37" with aeration.
Viability determinations. The direct determination of the percentage viable
bacteria in a suspension by a slide culture method (Postgate, Crumpton & Hunter,
1961) and counts of viable bacteria, were made as previously described (Strange
et al. 1961).
Chilling. Bacteria were usually separated from the culture by centrifugation and
resuspended at a suitable concentration in the same diluent as that in which they
were subsequently chilled. With experiments shown in Figs. 1 and 3, culture
directly from the growth vessel was used. Rapid cooling of bacterial suspensions
was achieved by dilution (1/50-1/100) in cold diluent held in a temperaturecontrolled bath containing aqueous ethylene glycol at 0". Concentrated bacterial
suspensions were initially cooled to near 0" by contact with brine at -10" before
placing in the bath. When freezing occurred during cooling, the suspension was discarded. The diluents used were buffered saline which contained: NaCl(O.13~)and
appropriate concentrations of K,HPO, + KH2P0, (0-02M PO,) to give the required
pH value; 0-05-0-15M 2-amino-2(hydroxymethyl)-l:3-propanediol(tris) +HC1 to
give the required pH value (tris buffer); distilled water. All diluents and diluents +
additives were filtered through a well-washed filter membrane before use.
Materials. Distilled water was passed through a mixed-bed ion-exchange resin
(Amberlite MB-1 from British Drug Houses Ltd.) column before use. Whenever
possible, Analytical Reagent Grade substances were used, Hydrated disodium
adenosine-5-triphosphate(ATP) was obtained from Sigma Chemical Co., St Louis,
Missouri, U.S.A.; dehydrated firefly tails were obtained from L. Light & Co. Ltd.;
Oxoid filter membranes, grade A.P., from 0x0 Ltd.
Analytical Methods. The total amino acid content of samples was determined by a
ninhydrin colorimetric method (Yemm & Cocking, 1955) with alanine as the standard. Amino acids were identified by %dimensional paper chromatography with
phenol saturated with water in an atmosphere of NH, as first solvent, butanol+
acetic acid +water (40 10 50, by vol. ; upper phase) as second solvent and ninhydrin as the spray reagent. ATP was determined by the following firefly luminescence technique which is essentially that described by Strehler & Totter (1952):
reaction mixtures (4 ml.) contained 0.02 M sodium arsenate buffer (pH 7*4),3mMMga+, 0-0.2pg. ATP and enzyme extracted from 2.5 mg. firefly tails; at the addition of enzyme, a stop-watch was started and luminescence was measured for
0.5 min. (beginning 1min. after mixing) with a scintillation counter constructed by
+ +
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
Effect of chibbing on Aerobacter aerogenes
721.
our colleague Mr S. Lovett. During a series of ATP determinations, counts were
obtained for appropriate blank and standard mixtures at regular intervals. Mag
nesium was determined by the Eriochrome Black T colorimetric procedure (Levine
& Cummings, 1956) and calcium was detected by precipitation as the oxalate.
Ultraviolet (u.v.) absorption was measured in a Unicam quartz spectrophotometer,
model S.P. 500,with a 1 cm. light path. Bacterial dry weights were determined as
previously described (Strange et al. 1961).
RESULTS
InJlueizce of bacterial growth phase o n the lethal effect of chilling
Figure 1 shows a typical growth curve for Aerobacter aerogenes in defined medium
and the sensitivity to chilling of bacteria harvested at intervals during the growth
period. Exponential-phase organisms were most sensitive to chilling after growth
for about 132 min. and then became progressively less sensitive up to 168 min.
before division ceased. Differences in the sensitivity of Escherichia culi to chilling
during exponential growth were reported by Hegarty & Weeks (1940) and Meynell
(1958; Fig. 1). Loss of viability of susceptible bacterial suspensions held at 0'
was progressive and bacteria removed immediately after chilling were completely
viable. Figure 2 shows survival curves for bacteria harvested a t four different times
during the growth period and then held at 0". The susceptibility to chilling of bacteria
in different stages of the division cycle was examined by means of a partially
synchronized culture (Fig. 3). During the initial lag phase, the bacteria became
almost completely resistant to chilling but sensitivity increased immediately
division began. In this and other experiments resistance to chilling during the
first division lag period (i.e. when organisms were increasing in size before the second
division) did not increase to that of stationary phase or initial lag-phase bacteria,
but remained unchanged or increased only slightly before decreasing during the
second division (Fig. 3). It is of interest that, whereas freshly harvested stationaryphase organisms were relatively resistant to chilling, after storage for 20 hr. at 37"
in buffered saline with aeration their resistance decreased and was little higher than
that of exponential-phase bacteria (Fig. 3, t = 0).
Eflect of the diluelzt
In the experiments above the ionic strength of buffered saline used as a diluent
was similar to that of 'physiological saline' (NaC1, 0.9 yo, w/v). When similar
concentrations of the same batch of exponential-phase organisms were chilled in
this and other diluents, the losses of viability shown in Table 1 were obtained. In
several experiments, losses were higher in tris buffer than in the other diluents. The
high loss of viability occurring in this diluent during chilling was not due to contamination of tris salt by heavy metals since the addition of disodium ethylenediamine tetraacetic acid (EDTA; 0 . 3 2 m ~to
) tris buffer had no effect on the results.
Of these diluents, distilled water was usually the least toxic but with a few batches
of organisms a slightly greater loss of viability occurred on chilling in water than in
buffered saline.
Loss of viability of bacterial suspensions held at 0" occurred progressively with
time (Fig. 2) and it was of interest to determine whether sudden warming after a
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
R. E. STRANGE
AND F. A. DARK
722
period at 0" decreased the rate of viability loss. A suspension of washed exponentialphase organisms in buffered saline (pH 6 - 5 ) was cooled to 0' and samples, removed
immediately and at intervals, were rapidly warmed to 18" in a water bath. Chilling
f y v
5 s 04
(a)
E c0.2
>>
E-
.E
M
(4
Time (min.)
Fig. 1
0
15
30
45
Time (min.)
60
Time (min.)
Fig. 2
Pig. 3
Fig. 1. Effect of growth phase on susceptibility of Aerobacler aerogenes to cold shock.
(a) Growth curve in carbon-limiting mannitol+ salts medium. (b) Viabilities of samples
30 min. after dilution to about lo8bacterialml. in buffered saline held a t 20". (c) Viabilities
after a similar period a t 0".
Fig. 2. Survival of exponential and stationary phase Aerobacter aerogenes a t 0". Bacteria
were separated from the culture by centrifugation and diluted to about 108/ml. in
buffered saline a t 0". Exponential phase organisms (0)
; bacteria in the stationary
phase for 15 min. (a),75 min. ( x ), and 135 min.
(a).
Fig. 3. Susceptibility of Aerobacter aerogenes to cold shock during partially synchronous
growth. (a) Growth curve in defined carbon-limiting medium determined by viable
counts ( 0 )and turbidity ( x ). (b) Viabilities of samples 30 min. after dilution to about
108 bacterialml. in buffered saline a t 20". (c) Viabilities after a similar period a t 0".
followed immediately by warming had no effect on viability, whereas after chillingfor
5 min. or more, suspensions continued to lose viability a t 18' but a t a slower rate
than a t 0' (Fig. 4).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
Effect of chilling on Aerobacter aerogenes
723
Influence of bacterial concentration
Five suspensions containing different concentrations of a batch of exponentialphase organisms in buffered saline were held at 0" for 1hr. Survival curves showed
that the death rate was greater the sparser the population (Fig. 5 ) . A similar but
less pronounced phenomenon occurred in tris buffer: suspensions of 5 x 107, 109 and
1010 viable bacteria/ml. tris buffer had viabilities of 8, 57 and 96% respectively,
immediately after chilling, and 8, 5 and 53% respectively, after 1 hr. at 0".
In the case of concentrated exponential-phase suspensions, it seemed possible
that the relatively great resistance to chilling was due to the appearance in the
suspending fluid of material with protective activity. To examine this possibility,
Table 1. Efect of the diluent on the loss of viability of exponential phase
Aerobacter aerogenes held at 0
' and 20"
Bacteria separated from the culture by centrifugation were diluted to about lO*/ml.
h diluents held at 0" and 20".
Experiment
no.
1
2
3
4
yo viability of suspension after
r
Diluent
Buffered saline, pH 6-5
0.05 M-trisbuffer, pH 7.5
Distilled water
( pH 6.0
PH 6-5
pH 7.5
0.05~-trisbuffer, pH 7.5
0.05~-trisbuffer, pH 7.5
+ 0.32M-EDTA
Distilled water
Buffered saline, pH 6.5
O.O5~-trisbuffer, pH 7.5
0-1Oiw-trisbuffer, pH 7.5
O.lS~-trisbuffer, pH 7.5
Distilled water
Buffered saline, pH 6.5
0-05iw-tris buffer, pH 7.5
Distilled water
45
h
min. at 0" 45 min. at 20"
32
1
75
54
37
42
21
1
>
98
98
98
99
99
1
56
50
28
15
25
85
8
<1
94
99
98
97
97
86
98
99
96
supernatant liquid separated from a chilled (1hr., 0")suspension initially containing
1.3x 1O1O viable exponential-phasebacteria/ml. buffered saline was passed through a
filter membrane and tested as follows : tubes containing (a)filtrate, (b) filtrate +
3 vol. buffered saline, ( c ) buffered saline, were cooled and sufficient fresh suspension
was added to give about 108 organisms/ml. each diluent. After 45 min. at O", the
viabilities of ( a ) ,(b) and ( c ) were 99, 94 and 23 yo,respectively. When filtrates from
chilled (1hr., Oo) exponential- and stationary-phase bacterial suspensions of about
the same concentration (1010 bacteria/ml.) were tested in a similar manner, viabilities after 45 min. at 0" were 98 and 64 yo, respectively, as compared with 24 yo
in the buffered saline control. The activity of protective material present in filtrates
was not affected by heating for 15min. at 100". These and other experiments
showed that filtrate from chilled exponential-phase bacteria and, to a lesser extent,
G. Microb.
46
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
XXIX
R.E. STRANGE
AND F. A. DARK
724
from chilled stationary-phase organisms, contained substances which protected
exponential-phase Aerobacter aerogenes from the lethal effect of chilling.
Examination of leakage products from chilled bacteria
The identification of the protective material present in filtrates from chilled
bacterial suspensions was complicated by the fact that the time taken to prepare
the relatively concentrated suspensions required for analytical investigation was
x
t
0
30
-
-L
15
30
45
Time (min.)
60
75
0
I
10
1
20
Fig. 4
1
30
1
40
Time (min.)
1
1
50
60
Fig. 5
Fig. 4. Survival of exponential phase Aerobacter aerogenes chilled a t 0" for short periods
and then held a t 1
8
'
. Bacterial suspensions were prepared as in Fig. 2. Viability at 18'
immediately after chilling ( ), after chilling for 5 min. ( x ) and.after chilling for 10 min.
( A ) ; viability of control suspension held throughout a t 0' ( 0 ) .
Fig. 5. Effect of population density on the survival of exponential phase Aerobacter
aerogenes at 0". Bacteria, separated from the culture by centrifugation, were resuspended at different concentrations in buffered saline held a t Oo. Viabilities immediately
and a t intervals after chilhg of suspensions hitially containing: 6 x lo7( A)y1.2 x 108(A),
1.2x 109 ( x )y 2.4 x 109 (a),1.2x 1010 (0)
viable bacteria/ml.
sufficient to affect the sensitivity of bacteria to chilling. For example, a suspension
diluted and chilled 10 min. after the bacteria were harvested lost 60% viability in
45 min. at 0' as compared with 10 yo when the suspension was held for 45 min. at
20' before chilling for the same period. The problem could not be resolved by direct
chilling of the culture because medium constituents and growth products interfered
with subsequent analysis. In the experiments described, suspensions were chilled
15-20 min. after harvesting the organisms. Bacteria, separated from an exponentialphase culture by filtration through filter membrane, were washed several times on the
0~
pad with buffered saline and resuspended at a concentration of 3 - 2 ~ 1viable
bacterialml. (99 yo viable). Part of the suspension was rapidly cooled to 0' and the
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
EJffect of- chiZEing on Aerobacter aerogenes
725
remainder held at 20'. Samples were taken immediately and at intervals from each
suspension for viability determinations and analyses. The latter samples were
immediately freed from bacteria by filtration (filter membrane) and the filtrates
analysed for ninhydrin-reacting substances and ATP. The concentration of ninhydrin-reacting substances in filtrates was expressed as percentage total amino
acids (as alanine) extractable with cold O-~N-HC~O,
from an equal volume of whole
bacterial suspension. Cooled suspension (4 ml.) was acidified by the addition of
HCIO, (72 %; 0.17 ml.) and left for 30 min. a t 0'. After centrifugation, the supernatant fluid was neutralized with ~ N - K O Hfiltered
,
through a membrane filter and
analysed for amino acids. Total amino acids (as alanine) in exponential phase
Aerobacter aerogernes suspensions accounted for about 1 % of the bacterial dry weight.
The results (Fig. 6) showed that the chilled suspension progressively lost viability,
whereas the unchilled suspension remained completely viable, and the leakage of
ninhydrin-reacting substances and also of ATP was greater from chilled bacteria.
Paper chromatographic examination of similar filtrates obtained from more concentrated bacterial suspensions showed that the ninhydrin-reacting material
could be largely accounted for as free amino acids and peptides of relatively small
molecular weight. After electrolytic desalting and concentration, filtrate from a
chilled suspension (1 hr., 0') gave spots corresponding in position to aspartic acid,
glutamic acid, glycine, serine, alanine, methionine, valine, leucine, arginine and
histidine, with at least two unidentified components; acid hydrolysis of the filtrate
( ~ N - H C20
~ , hr., 106') caused an increase in intensity of several spots, the disappearance of two unidentified constituents and the appearance of several other
common amino acid spots. Paper chromatograms of filtrate from suspensions of
similar concentration held for the same period at 20' showed the presence of
several amino acids but in much lower concentrations than those found with
chilled suspensions. Neither acid-insoluble protein nor ribonucleic acid (RNA)
were detected as constituents of the material which leaked from chilled bacteria.
For example, filtrate from a chilled exponential-phase suspension of organisms
equivalent to 26 mg. dry weightlml. (about 5 x 1010 bacterialml.) gave no turbidity
with HCIO, ( 0 . 5 ~under
)
conditions where protein or RNA equivalent to 0.04 yo
bacterial dry weight would have been detected. Although acid-insoluble RNA was
absent, filtrates from chilled and unchilled bacterial suspensions contained u.v.absorbing substances in greater concentration than could be accounted for by ATP.
Maximum absorption occurred at a wavelength near 260 mp and the initial rate of
leakage of these substances was much greater from chilled than from unchilled
organisms (Fig. 7). However, on storage of suspensions for longer periods at 20' the
u.v.-absorption of the suspending liquid increased steadily, whereas a t 0' it reached
a maximum after about 30 min. and then remained unchanged. This was due to
metabolism of endogenous RNA which occurs in starved Aerobacter aerogenes SUSpensions held at 20', resulting in the excretion of nucleic acid bases (Strange et al.
1961), whereas at Oo, negligible degradation of RNA was found to occur. Neither
magnesium nor calcium ions were found in protective filtrates under conditions
where 0 . 5 m of
~ either would have been detected.
46-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
726
R. E. STRANGEAND 3. A. DARK
Protective eflect of ezogenous substances
The possibility that the protective effect of leakage products from chilled organisms was due to the presence of ATP or amino acids was investigated. Also, the
influence during cold shock of various other substances, not detected in leakage
products, was examined.
ATP. The addition of ATP at a concentration similar to that found in protective
filtrates ( 1 0 - 2 0 p ~ )to buffered saline or tris buffer diluents afforded no protection to
exponential phase organisms against cold shock,
0
2s
Time (min.)
Tlme (min.)
Fig. 6
Fig. 7
45
Fig. 6. Leakage of endogenous constituents from exponential phase Aerobacter aerogenes
held in suspension at 0
' and 20'. Washed organisms were resuspended in buffered saline
(equiv. 2.1 mg. baderial dry weightlml.). (a)Loss of viability at 0' (0)and 20' ( 0 ) .
(b) Concentration of ninhydrin-reacting substances as alanine (percentage total cold
acid-extradable amino acids in whole suspension) in filtrate from suspension at 0' ( 0 )
and 20' ( 0 ) ;ATP (pg./ml.) in filtrate from suspension at '
0 ( A ) and at 20' (A).
Fig. 7. Leakage of u.v.-absorbing substances from exponential Ambacter aerogenes held
in suspension at 0" and 20'. Bacterial suspension (equiv. 2.6 mg. bacterial dry weight/
ml.) was prepared as in Fig. 6. Spectrophotometricmeasurements at 260 mp on filtrate
from suspension a t 0' ( 0 )and 20' ( 0).
Amiw acids. Loss of viability of bacteria chilled in buffered saline containing the
10 amino acids detected in leakage products (each 0 . 5 m ~occurred
)
at a slower rate
than in buffered saline alone (Fig. 8 ) . The addition of 7 other amino acids (each
0 . 5 m ~to
) this mixture did not improve the protection afforded and neither mixture
was as effective as leakage products (Fig. 8). The influence of the 17 amino acid
mixture was similar after fivefold dilution with buffered saline, but below this
concentration the protective effect disappeared.
Divalent metal ions. The leakage of endogenous constituents which occurred as a
result of cold shock indicated that a permeability control mechanism was affected by
chilling; such a mechanism would presumably be located in the cytoplasmic membrane. The known stabilizing effect of magnesium ions on isolated protoplast
membranes (Weibull, 1956) and spheroplasts (Lederberg, 1956 ; McQuillen, 1958)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
727
Effect of chiZZhg on Aerobacter aerogenes
suggested that metal ions might protect bacteria during cold shock. The effect of
adding magnesium and other divalent metals to the diluent in which bacteria were
chilled was investigated. Bacteria suspended in buffered saline or t r i s buffer were
substantially protected during chilling in the presence of Mg2+, Ca2+ or Mn2+
(Table 2; Fig. 9). With distilled water as the diluent, Mn2+ was slightly toxic,
Ca2+ or Mg2+ had little effect and Co2+, Zn2+ or Fe2+ were very toxic (Table 2). The
pH value of suspensions in distilled water was decreased by the presence of metals;
this may have contributed to the loss of viability which occurred.
Sucrose and erythritol. Meynell(l958) showed that exponential-phase Escherichia
coli suspensions were protected during chilling by sucrose (0.3M). Chilled Aerobacter
Table 2. Eflect of divalent metal ions on the loss of viability of ezportential-phase
Aerobacter aerogenes suspensions at 0
'
Bacteria, separated from the culture by centrifugation, were diluted to about 108/ml.
in various diluents.
1
2
Diluent
buffered saline (pH 6.5)
suspension
after 45 min.
at 0"
Addition to diluent
nil
MgSO,: 0.5, 1.0,5 . 0 1 n ~
*MnSO,: 0.5, 1.0,5 0 m ~
CaC1,: 0.5, 1.0, 5 . 0 ~
nil
10
19, 16, 53
7, 8, 84
10, 16, 71
<1
94
0-05~x-trisbuffer (pH 7 - 5 )
74
88
nil
3
-
yo viability of
Bacteria
batch
no.
distilled water
*
MgSO,: 1.0, 2-5, 5 . 0 1 n ~
MnSO,: 1.0,2.5, 5 - 0 m ~
CaCI,: 1.0,2-5, 5 - 0 m
COSO,: 1.0,2.5, 5 . 0 m ~
ZnSO,: 1.0,2-5, 5-Ormur
FeSO,: 1.0, 2.5, 5 - 0 m ~
85
76,79, 83
28, 37, 57
88,81, 91
<1
<1
<1
Precipitation occurred when Mn2+ was added to this diluent.
wogenes suspensions were also protected by sucrose (Fig. 10). Analysis of filtrates
of chilled bacterial suspensions in buffered saline with and without sucrose (0.3~)
showed that the concentration of leakage products was considerably less in the
presence of sucrose. This sugar does not penetrate into A. aerogenes whereas
I-erythritol does (Postgate & Hunter, 1961). When bacteria were chilled in buffered
saline, buffered saline + sucrose ( 0 . 3 ~ )and buffered saline +erythritol (0.3M),
losses of viability after 1 hr. at Oo were 99, 24 and SS%, respectively.
Spermine. Since spermine ( 1 0 - 3 ~ ) stabilizesPastezlrella tularemis and Escherichia
coli spheroplasts against osmotic damage (Mager, 1959)it was possible that this substance might protect bacteria during cold shock, At Oo, the loss of viability of
exponential-phase Aerobacter aerogerm in buffered saline containing ~ O - * M spermhe
was more rapid than in buffered saline alone. In the presence of 1 O m 6 spermbe
~
there was a transient protective effect: losses of viability after 10,20 and 80 min. at
46-3
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
R.E. STRANGEAND F. A. DARK
728
0' were 34, 65 and 75 %, respectively, as compared with 78, 82 and 86 %, respectively, in the control suspension.
Revival of chilled bacteria
Protective substances which appeared to exert their influence on bacteria during
chilling may have in fact acted after chilling, by reviving moribund bacteria so
that they were able to grow on nutrient agar, When bacterial suspensions in buffered
saline were chilled and then amino acids, magnesium, calcium or manganese added,
viability was not affected. However, when leakage products from a chilled con100
80
h
-x
60
*-
5
m 40
5
20
0
10 20 30
Time (min.)
60
0
10 20 30
Time (min.)
60
-
..
Time (min.)
Fig. 8
Fig. 9
Fig. 10
Fig. 8. Influence of leakage products from chilled bacteria and amino acid mixtures on
the survival of exponential phase Aerobacter aerogms at 0'. Bacteria separated from
the culture were diluted to about 108/ml. in the solutions given held at 0".Viability in
filtrate (A),in amino acid mixture A (A),in amino acid mixture B ( 0 )and in buffered
saline (0).The filtrate was prepared from a suspension of chilled exponential phase
~
of the 10
organisms in buffered saline (1.3x 1010/ml.). Mixture A contained 0 - 5 m each
amino acids (L-isomers)found in leakage products (see text) dissolved in buffered saline;
mixture B contained amino acids in A + 0 . 5 m ~each of tyrosine, phenylalanine, lyshe,
a,e-diaminopimelic acid (meso), threonine (DL), cystine and tryptophan.
Fig. 9. Influenceof divalent metal ions on the survival of exponentialAerobacte~amogenes
at 0'. Viability of suspensions (about lo*bacteria/ml.) in tris buffer containing: 5 m ~ MgSO,( O), 5m~-CaC1,(A), 5m~-MnsO,(A);viability in tris buffer control ( 0 ) .
Fig. 10. Influence of sucrose on the survival of exponential phase Aerobacter aerogenes
at 0'. Bacteria separated from two separate cultures were diluted to about 108/ml. in
diluents a t 0'. Viability in buffered saline containing 0 . 3 sucrose
~
( 0 , A),in buffered
saline control ( 0 ,A).
centrated suspension were added to a dilute chilled suspension, the viability increased. A suspension of Aerobacter mogenes in buffered saline (8x 108 bacteria/ml.)
was held at 0' and samples, removed at intervals, were diluted with an equal
volume of either buffered saline or filtrate from chilled (1 hr., 0') exponential-phase
bacteria in buffered saline (3x 101O/ml.). Viabilities, determined immediately after
dilution with buffered saline, of bacteria held at 0" for 15, 30 and 60 min. were 39,
23 and 15 yo,respectively; after dilution with filtrate, the viabilities of the samples
'were 76, 72 and 35%, respectively. On chilling similar concentrations of these
bacteria in equal volumes of buffered saline and filtrate, viabilities after the same
periods at 0' were 98, 92, and 89%, respectively. Thus, the protective effect of
leakage products exerted during chilling was greater than the revival effect which
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
Eflect of chiZZing on Aerobacter aerogenes
729
these substances exerted on chilled bacteria. The revival effect was not due to
transfer of leakage products with bacteria onto slide cultures; in experiments where
the viability was determined by viable counts involving a 1/105dilution of the sample
before plating, an increase in the viability of a chilled suspension occurred after the
addition of leakage products.
DISCUSSION
The effect of the growth phase on the susceptibility of dilute Aerobacter aerogenes
suspensions to cold shock was similar to that found with Escherichia coli (Sherman
& Albus, 1923; Sherman & Cameron, 1934; Hegarty & Weeks, 1940; Meynell, 1958)
and Pseudomonas pyocyanea (Gorrill & McNeil, 1960). Also, as previously found,
the lethal effect of cold shock depended on the diluent. However, with exponentialphase A . aerogenes the lethal effect was less in distilled water than in buffered saline
or tris buffer, whereas with growing P.pyocyanea it was greater in distilled water
than in Ringer’s solution, buffer solution or saline (Gorrill & McNeil, 1960). In this
connexion, it is of interest that the death rate of steadily growing A . aerogenes
caused by freezing in liquid nitrogen was less with suspensions in distilled water than
with those in saline phosphate or saline tris buffers (Postgate & Hunter, 1961).
The finding that a rapid leakage of endogenous constituents occurs on chilling
bacteria supports the view that the lethal effect of cold shock is due to interference
with a bacterial permeability control mechanism (Meynell, 1958). The population
density phenomenon, and the protective effect that filtrates of dense populations
exert on the survival of sparser populations, together imply that one or more of these
leakage products is necessary for their survival and can be resorbed by the bacteria
from their external environment. This resorption must be a rapid process, since
brief exposure of moribund populations in buffered saline to leakage products from
a denser population increased their viability. This reactivation phenomenon was not
only observed when viability was determined by slide culture but also in experiments where viability was determined by a plate counting technique involving a
1/105 dilution of the suspension. The protective material clearly induced a permanent change in the population; one no longer influenced by dilution.
Our experiments show that during chilling the protective activity of leakage
products may be partially due to the presence of amino acids; ATP was also leaked
but it was inactive when added to diluents in which bacteria were chilled. Other
substances which protected bacteria but which were not detected or are unlikely
to be constituents of leakage products are calcium, magnesium, manganese and
sucrose. The mechanisms by which these substances protect during cold shock are
probably different from the one which operates with leakage products or amino
acids. The known stabilizing effect of magnesium on isolated protoplast membranes
(Weibull, 1956) and spheroplasts (Lederberg, 1956 ; McQuillen, 1958) suggests that
the metal ions decrease the permeability of the cytoplasmic membrane in the intact
cell during chilling. The protective influence of sucrose shown previously with
Escherichia coli by Meynell (1958) is probably due to its osmotic activity which
causes a decrease in the rate of diffusion of constituents out of bacteria. Erythritol
in equivalent concentration did not have a similar protective effect, presumably
because this substance is osmotically neutral with Aerobacter aerogenes (Postgate &
Hunter, 1961).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25
730
R. E. STRANGEAND I?. A. DARK
Despiteits reported stabilizingeffectonEscherichiacoli spheroplastsagainst osmotic
shock (Mager, 1959),low concentrations of spermine had only a transient protective
effect on chilled Aerobacter aerogenes. However, higher concentrations of spermine
were toxic to chilled organisms and the losses of viability which occurred with
lower concentrations of the substance may have been due to this toxic effect as well
as to cold shock.
We thank Dr J. R. Postgate for helpful criticism and suggestions, Mr A. G. Ness for
some of the analyses, Mr S. Lovett for the use of a scintillation counter and Mr B.
Phillips for technical assistance.
REFERENCES
GORRILL,R. H. & MCNEIL,E. M. (1960). The effect of cold diluent on the viable count of
Pseudomonas pyocganea. J . gen. Microbwl. 22,437.
HEGARTY,
C. P. & WEEKS,0. B. (1940). Sensitivity of Escherichia coti to cold-shock
during the logarithmic growth phase. J. Bact. 39,475.
LEDERBERG,
J, (1956). Bacterial protoplasts induced by penicillin. Proc. nat. Acad. Sei.,
Wash. 42, 574.
LEVINE, R. M. & CUMMINGS,J. R. (1956). A sensitive, accurate spectrophotometric
method for the determination of magnesium in serum. J. EoZ. Chena. 221, 735.
MCNAIR
SCOTT,
D. B. & CHU, E. (1958). Synchronised division of growing cultures of
Escherichia coli. E z p . Cell Res. 14,166.
MCQUILLEN,K.(1958). Lysis resulting from metabolic disturbances. J.gen. Microbiol. 18,
498.
MAGER,
J. (1959). The stabilising effect of spermine and related polyamines on bacterial
protoplasts. Biochim. biophys. acta, 36, 529.
MEYNELL,
G.G.(1958). The effect of sudden chilling on Escherichia coli. J . gen. Mimobiol.
19,380.
POSTGATE,
J.R.,CRUMPTON,J. E. & HUNTER,
J. R. (1961).The measurement of bacterial
viabilities by slide culture. J. gen. Microbiol. 24, 15.
POSTGATE,
J. R. & HUNTER,
J. R. (1961). On the survivd of frozen bacteria. J . gen.
Microbiol. 26, 367.
SHERMAN,
J. M. & ALBUS,W.R.(1923). Physiological Youth in bacteria. J . Bact. 8, 127.
SHERMAN,
J. M. & CAMERON,
G. M. (1934). Lethal environmental factors within the
natural range of growth. J . Bact. 27, 341.
STRANGE,
R. E.,DARK,F. A. & NESS,A. G. (1961). The survival of stationary phase
Aerobacter wogenes stored in aqueous suspension. J . gm. Mierobiol. 25,61.
STREELER,
B. L. & TOTTER,
J. R. (1952). Firefly luminescence in the study of energy
transfer mechanisms. I. Substrate and enzyme determination. Arch. Biochem. Biophys.
40, 28.
WEIBULL,
C. (1956). Bacterial protoplasts; their formation and characteristics. In Bacterial Anatomy, Symp. SOC.gen. Microbiol. 6, 111.
YEMM,
E. W. & COCKING,
E. C. (1855).The determination of amino-acids with ninhydrin.
Analyst, 80, 209.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 16 Jun 2017 04:12:25